v3.18.1

JayDDee · Oct 11, 2021 · 47cc5dc · 47cc5dc
1 parent 2cd1507
commit 47cc5dc
Show file tree

Hide file tree

Showing 14 changed files with 2,013 additions and 2,783 deletions.
diff --git a/INSTALL_LINUX b/INSTALL_LINUX
@@ -32,14 +32,26 @@ but different package names.
 $ sudo apt-get install build-essential automake libssl-dev libcurl4-openssl-dev libjansson-dev libgmp-dev zlib1g-dev git
 
 SHA support on AMD Ryzen CPUs requires gcc version 5 or higher and
-openssl 1.1.0e or higher. Add one of the following to CFLAGS for SHA
-support depending on your CPU and compiler version:
+openssl 1.1.0e or higher.
 
-"-march=native" is always the best choice
+znver1 and znver2 should be recognized on most recent version of GCC and
+znver3 is expected with GCC 11. GCC 11 also includes rocketlake support.
+In the meantime here are some suggestions to compile with new CPUs:
 
-"-march=znver1" for Ryzen 1000 & 2000 series, znver2 for 3000.
+"-march=native" is usually the best choice, used by build.sh.
 
-"-msha"  Add SHA to other tuning options
+"-march=znver2 -mvaes" can be used for Ryzen 5000 if znver3 is not recongized.
+
+"-mcascadelake -msha" or
+"-mcometlake -mavx512 -msha" can be used for Rocket Lake.
+
+Features can also be added individually:
+
+"-msha" adds support for HW accelerated sha256.
+
+"-mavx512" adds support for 512 bit vectors
+
+"-mvaes" add support for parallel AES
 
 Additional instructions for static compilalation can be found here:
 https://lxadm.com/Static_compilation_of_cpuminer

diff --git a/RELEASE_NOTES b/RELEASE_NOTES
@@ -65,10 +65,24 @@ If not what makes it happen or not happen?
 Change Log
 ----------
 
+v3.18.1
+
+More speed for scrypt:
+ - additional scryptn2 optimizations for all CPU architectures,
+ - AVX2 is now used by default on CPUS with SHA but not AVX512,
+ - scrypt:1024 performance lost in v3.18.0 is restored,
+ - AVX512 & AVX2 improvements to scrypt:1024.
+
+Big speedup for SwiFFTx AVX2 & SSE4.1: x22i +55%, x25x +22%.
+
+Issue #337: fixed a problem that could display negative stats values in the
+first summary report if the report was forced prematurely due to a stratum
+diff change. The stats will still be invalid but should display zeros.
+
 v3.18.0
 
 Complete rewrite of Scrypt code, optimized for large N factor (scryptn2):
-  - AVX512 & SHA support for SHA256, AVX512 has priority,
+  - AVX512 & SHA support for sha256, AVX512 has priority,
   - up to 50% increase in hashrate,
   - memory requirements reduced 30-60% depending on CPU architecture,
   - memory usage displayed at startup,